Waveform Quantization and Compression - Digital Signal Processing: Fundamentals and Applications - page 537

% Note: the midtread method is used in this quantizer.
%

delta ¼ 2*abs(Xmax)/(2^NoBits-1);
pq ¼ indx*delta;
Program 11.7. MATLAB function for calculation of signal to quantization noise ratio (SNR).

function snr ¼ calcsnr(speech, qspeech)
% function snr

calcsnr(speech, qspeech)
% This routine was created for calculation of SNR.
%
% speech: original speech waveform.
% qspeech: quantized speech.
% snr: output SNR in dB.
%
% Note: the midrise method is used in this quantizer.
%

¼

qerr ¼ speech-qspeech;
snr ¼ 10*log10(sum(speech.*speech)/sum(qerr.*qerr))
Program 11.8. Main program for digital m -law encoding and decoding.

load we12b.dat
for i ¼ 1:1:length(we12b)
code8b(i) ¼ dmuenc(12, we12b(i));

% Encoding

qwe12b(i) ¼ dmudec(code8b(i)); % Decoding

end
subplot(4,1,1),plot(we12b);grid
ylabel( ' a ' );axis([0 length(we12b) -1024 1024]);
subplot(4,1,2),plot(code8b);grid
ylabel( ' b ' );axis([0 length(we12b) -128 128]);
subplot(4,1,3),plot(qwe12b);grid
ylabel( ' c ' );axis([0 length(we12b) -1024 1024]);
subplot(4,1,4),plot(qwe12b-we12b);grid
ylabel(

);axis([0 length(we12b) -40 40]);
Program 11.9. The digital m -law compressor.

d );xlabel( Sample number

' ' ' '

function [cmp_code ] dmuenc(NoBits, value)

% This routine is created for simulation of 12-bit mu law compression.
% function [cmp_code ] ¼ dmuenc(NoBits, value)
% NoBits ¼ number of bits for the data
% value ¼ input value
% cmp_code ¼ output code
%

¼

scale ¼ NoBits-12;
value ¼ value*2^(-scale); % Scale to 12 bit
if (abs(value) > ¼ 0) & (abs(value)<16)
cmp_code ¼ value;
end
if (abs(value) > ¼ 16) & (abs(value)<32)